diff --git a/bixbench/graders.py b/bixbench/graders.py
index f2e81b0..ddf9682 100644
--- a/bixbench/graders.py
+++ b/bixbench/graders.py
@@ -159,8 +159,7 @@ def _grade_range_verifier(
     ) -> GradeResult:
         lower, upper = ast.literal_eval(target)
         correct = lower <= float(predicted) <= upper
-        if correct:
-            grade_type = GradeType.CORRECT
+        grade_type = GradeType.CORRECT if correct else GradeType.INCORRECT
 
         return GradeResult(
             grade=grade_type.numeric_grade,
diff --git a/pyproject.toml b/pyproject.toml
index a941145..697cd98 100644
--- a/pyproject.toml
+++ b/pyproject.toml
@@ -56,6 +56,9 @@ check-hidden = true
 ignore-words-list = "coefficent,LasR"
 skip = "uv.lock,bixbench_results/baseline_eval_data/*,bixbench-v1.5_results/*"
 
+[tool.pytest.ini_options]
+pythonpath = ["."]
+
 [tool.refurb]
 enable_all = true
 ignore = [
diff --git a/tests/test_utils.py b/tests/test_utils.py
index ab30405..482803e 100644
--- a/tests/test_utils.py
+++ b/tests/test_utils.py
@@ -109,6 +109,28 @@ async def test_grade_mcq_answer(
     assert grade_result.refusal == expected_refusal
 
 
+@pytest.mark.asyncio
+@pytest.mark.parametrize(
+    ("target", "predicted", "expected_grade", "expected_correct"),
+    [
+        pytest.param("(1.0, 3.0)", "1.0", 1, True, id="lower_bound"),
+        pytest.param("(1.0, 3.0)", "2.0", 1, True, id="inside_range"),
+        pytest.param("(1.0, 3.0)", "3.0", 1, True, id="upper_bound"),
+        pytest.param("(1.0, 3.0)", "4.0", 0, False, id="outside_range"),
+    ],
+)
+async def test_grade_range_answer(
+    target: str, predicted: str, expected_grade: int, expected_correct: bool
+):
+    grade_result = await MCQGrader().grade(
+        target, predicted, evaluation_mode="range_verifier"
+    )
+
+    assert grade_result.grade == expected_grade
+    assert grade_result.correct is expected_correct
+    assert grade_result.refusal is False
+
+
 @pytest.mark.parametrize(
     ("grades", "is_refused", "metrics"),
     [